Envibe runs on your own machine — the client starts it, so there is no endpoint to ping. 49 installs a week from npm. Last commit 3 Feb 2026.
Secure permission layer between AI agents and your .env files
Today is the operative word: we check Envibe every 15 minutes and re-read its code on every release. Watch it and you find out the day that stops being true.
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add envibe -- npx -y envibe-mcp
{
"mcpServers": {
"envibe": {
"command": "npx",
"args": [
"-y",
"envibe-mcp"
]
}
}
}
[mcp_servers.envibe]
command = "npx"
args = ["-y", "envibe-mcp"]
{
"mcpServers": {
"envibe": {
"command": "npx",
"args": [
"-y",
"envibe-mcp"
]
}
}
}
{
"servers": {
"envibe": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"envibe-mcp"
]
}
}
}
Large files and secrets between people and agents, never through the chat or the model.
Test your AI agents for security and safety weaknesses, and re-verify the fixes.
Ask about your Kodem Security data and act on findings in plain language, with your permissions.
The deterministic attack-surface perception layer for AI security agents
Send encrypted files and secrets, collect files and get documents signed, from your AI.
Deterministic, zero-token security scanner your AI agent calls to find and re-verify issues.
Pre-execution security for AI agents on Base: guard token/tx/address + per-agent spend firewall.
Let AI agents securely read the text messages on your phone.
Answers built from our own checks of this server.